Hacker News: larsrc

New comment by larsrc in "Nepenthes is a tarpit to catch AI web crawlers"

larsrc — Thu, 16 Jan 2025 20:44:36 +0000

I've been considering setting up "ConfuseAIpedia" in a similar manner using sentence templates and a large set of filler words. Obviously with a warning for humans. I would set it up with an appropriate robots.txt blocking crawlers so only unethical crawlers would read it. I wouldn't try to tarpit beyond protecting my own server, as confusion rogue AI scrapers is more interesting than slowing them down a bit.

New comment by larsrc in "100x defect tolerance: How we solved the yield problem"

larsrc — Thu, 16 Jan 2025 12:03:51 +0000

How do these much smaller cores compare in computing power to the bigger ones? They seem to implicitly claim that a core is a core is a core, but surely one gets something extra out of the much bigger one?

New comment by larsrc in "Behind the Scenes: Alien Worlds – Jumpship over a tidally locked planet"

larsrc — Tue, 31 Dec 2024 09:48:07 +0000

Cool to see the whole process behind it. Some sneaky tricks there, and much knowledge. Unfortunately, I find the picture doesn't show the tidally-lockedness very well. If it wasn't for the title, I would never have noticed. The haze on the sun side makes it look like clouds.

New comment by larsrc in "Cognitive load is what matters"

larsrc — Thu, 26 Dec 2024 14:42:08 +0000

Oh, and there's massive use of aspect-oriented programming, the least local paradigm ever!

New comment by larsrc in "Cognitive load is what matters"

larsrc — Thu, 26 Dec 2024 14:41:20 +0000

> If someone feels compelled to read every function either the functions are poor abstractions or the reader has trust issues, which may be warranted.

Or it's open source and the authors were very much into Use The Source, Luke!

New comment by larsrc in "Cognitive load is what matters"

larsrc — Thu, 26 Dec 2024 14:37:35 +0000

Some people are proud of making complex code. And too many people admire those who write complex code.

New comment by larsrc in "Cognitive load is what matters"

larsrc — Thu, 26 Dec 2024 14:34:24 +0000

Comments should describe what the type system can't. Connect, pitfalls, workarounds for bugs in other code, etc.

New comment by larsrc in "Cognitive load is what matters"

larsrc — Thu, 26 Dec 2024 14:32:12 +0000

But knowing what Brian was considering at the time is useful, both due avoiding redoing that and for realising that some constraints may have been lifted.

New comment by larsrc in "Why is it so hard to buy things that work well? (2022)"

larsrc — Mon, 16 Dec 2024 21:21:26 +0000

Even more infuriating is when the more expensive version is the same object but with extra unnecessary features added via software, to the detriment of usability.

New comment by larsrc in "Why is it so hard to buy things that work well? (2022)"

larsrc — Mon, 16 Dec 2024 21:02:25 +0000

And here I thought here would take about how new models come out so frequently that any information about long-term quality is long obsolete.

New comment by larsrc in "Training myself to run farther with Strava's API and an IoT dog feeder of M&Ms"

larsrc — Thu, 05 Dec 2024 17:41:37 +0000

I recently found that store-brand peanut M&M's (my favourite) were just as good, half the price, and fair-trade! Win-win!

When I joined my current job, I used M&M's as reward when learning the names and faces of the people there using memory cards. Also worked!

Fortunately I've found that Belegarth fighting is so much fun that I need no extra carrot.

New comment by larsrc in "Oncall shift should be Tuesday to Tuesday"

larsrc — Sat, 23 Nov 2024 15:39:42 +0000

+1! You can't travel very much, you can't go hiking or biking in places without cell coverage, your whatever thing you are busy with gets interrupted, you can get woken up in the middle of the night, etc etc. That deserves some compensation.

New comment by larsrc in "Against Best Practices"

larsrc — Mon, 18 Nov 2024 11:06:59 +0000

Follow best practices unless you can give a reason not to. "Best practice" is a shorthand for a lot of accumulated knowledge that you don't want to go over again every time. Also following BP makes the code more consistent and thus easier to understand. But when an argument arises, go back to the underpinnings of the best practice and work from there.

New comment by larsrc in "I Followed the Official AWS Amplify Guide and Was Charged $1,100"

larsrc — Thu, 14 Nov 2024 11:15:52 +0000

It's a bad user experience if you have to keep checking for hidden extra costs.

New comment by larsrc in "I Followed the Official AWS Amplify Guide and Was Charged $1,100"

larsrc — Thu, 14 Nov 2024 09:16:46 +0000

While I don't like Amazon one but, I appreciate that you don't go into the Evil Big Tech trope. Things at that scale are indeed complicated and hard to coordinate.

That said: fuck, that's expensive and poorly explained! Not doing anything cloud without hard limits!

New comment by larsrc in "TLA from First Principles"

larsrc — Tue, 05 Nov 2024 18:41:02 +0000

Wow, I totally missed that. Sorry.

New comment by larsrc in "Why software only moves forward"

larsrc — Tue, 05 Nov 2024 18:37:10 +0000

Moving software "backwards" is hard, but not impossible. There are ways to structure both data and code that make it easier. A change being "rollback-safe" can be an important question in designs and code reviews.

New comment by larsrc in "Cooking with black plastic is particularly crucial to avoid"

larsrc — Thu, 31 Oct 2024 10:09:25 +0000

I've been using wooden spatulas on my non-stick pans for years without any issues. Maybe only cheap non-sticks have trouble with them?

New comment by larsrc in "Cooking with black plastic is particularly crucial to avoid"

larsrc — Thu, 31 Oct 2024 10:06:48 +0000

And that's why you should pour boiling water over any implement that has been in contact with raw meat.

New comment by larsrc in "Cooking with black plastic is particularly crucial to avoid"

larsrc — Thu, 31 Oct 2024 10:04:19 +0000

And that even when using plenty of fat, a sticky pan can be a bother to clean.